EGG 11265-2035 UC-702

The Effect of Drought on Four Plant Communities in the Northern Mojave Desert.

Schultz, B. W. and W. K. Ostler, 1993. Plant Ecologist, Desert Research Institute, Reno, NV

Department Manager, EG&G Energy Measurements, La8 Vegas, NV.

Work supported by the U.S. Department of Energy, Nevada Operations Office under contract No. DE-AC08-93NVl.1265.

INTRODUCTION

Desert plant communities contain many perennial plant species that are well adapted to arid environments; therefore, one would intuitively believe that perennial desert species readily survive drought conditions. Abundant research on plant-soil-water relationships in North American deserts has shown that many species can maintain water uptake and growth when the soil-water potential is low. Little research, however, has focused on how prolonged drought conditions af€ect plant species i n vegetation associations in desert ecosystems. A prolonged and widespread drought occurred in much of the western United States, including the Northern Mojave Desert, from 1987 through 1991. During this drought period vegetation characterization studies, initiated in 1990, by the U. S. Department of Energy (DOE) at Yucca Mountain, Nevada, allowed EG&G Energy Measurements to collect data that could be used to infer how both desert vegetation associations and desert plant species reacted to a prolonged drought. This paper presents the preliminary results.

OBJECTIVES

1. To determine how vegetation associations in the Yucca Mountain area respond to a prolonged drought.

2. To determine if plant species that occur in two or more vegetation associations respond similarly to drought.

STUDY SITE DESCRIPTION

Four primary vegetation associations occur in the Yucca Mountain Project area: Larrea tridentata-Ambrosia dumosa, Larrea tridentata-Lycium andersonii-Grayia spinosa, Coleocrvne ramosissima, and Lvcium andersonii-Gravia spinosa (Figures la-le). The Coleosvne community consists of both low and high elevation (i.e., mountain summits versus valley bottoms) variations. Table 1 provides a relative description of each vegetation association, and the elevation and

I precipitation gradients that occur at Yucca Mountain.

METHODS

Twelve, 200 x 200-m ecological study plots (ESPs) were established in each vegetation association. The Coleosvne association had four ESPs in the upper

DlSCLAl M ER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

elevation association and eight ESPs in the lower elevation association.

Vegetation density measurements occurred in eight or ten, 2 x 50-m belt transects, located in each ESP. Each belt transect, in each ESP, was further divided into twenty-five, 2 x 2-m quadrats. Live and dead perennial plants, except seedlings from the current year, were identified by species in each 2 x 2-m quadrat. Data collection occurred in 1991 which was the first year that precipitation was sufficient to allow an accurate assess of living and dead plants.

The number of live and dead plants of each species in each ESP, in each vegetation association, was calculated. A ratio of live to dead plants (L:D) €or each species in each vegetation association was determined.

The L:D ratio was used as a common index to assess how species responded to prolonged drought across vegetation associations. Species that have L:D ratios well below 1:l are considered to have suffered substantial mortality from the drought. Species with L:D ratios well above 1:l are considered to have endured the drought well. Species with L:D ratios near 1:l are inconclusive.

An assumption was made that the L:D ratios prior to the drought were more or less equivalent among vegetation associations for individual species, and that changes in these ratios, between associations, were the result of drought conditions.

To test the hypothesis that the drought effected each of the vegetation associations similarly, we classified species that occurred in each vegetation association into two categories (L:D >1:1 and <l:l). A Chi-square test was conducted to assess if differences existed among vegetation associations.

RESULTS

Vesetation Association Response

The number of live and dead plants in each vegetation association are reported in Table 2. The percent of all live plants after the drought was 64. The low elevation Coleoqvne association had the lowest survival with 47.4%. The Larrea-Ambrosia and the Larrea-Lvcium-Gravia association had roughly the same survival with 58.1% and 58.9% respectively. The high elevation Coleoavne and the Lvcium-Gravia associations had much higher rates of survival at 79.2% and 82.0% respectively.

A Chi-square analysis performed on the 1ive:dead ratios showed that the drought did not have a similar effect on the vegetation associations at Yucca Mountain, Nevada (Table 3 ) . The Larrea-Ambrosia, high elevation Coleoqvne, and Lvcium-Gravia associations tended to have more species than expected in the >1:1 category, while the Larrea-Lvcium-Gravia and low elevation Coleoavne had more species than expected in the <1:1 category. The X of 28.0 (F.001, df=4) suggests that t h i s is not a random distribution.

2

Species Response

The response to drought of the 30 species analyzed is provided in Table 4. The percent of living plants for each species summed over all vegetation associations ranged from a low of 3% €or Orvzopsis hvmenoides to a high of 98% for Hilaria iamesii and Chrvsothamnus nauseosus. Only three species 0. hvmenoides, Stipa speciosa and Atriplex confertifolia had overall rates under 50%. The median value for all species was 77%.

Twelve of the thirty species analyzed occurred in all vegetation associations. Seven species occurred in only Lvcium-Gravia or the high elevation Coleouvne associations. Eight species occurred in all vegetation association except the high elevation Coleoavne. Two species occurred in all vegetation associations except the low elevation Coleoavne association and one species was absent from the high elevation Coleoavne and the Larrea-Ambrosia associations.

Species were present in most of the vegetation associations; however, they did not respond similarly across the vegetation associations. Species responses can be classified into three general categories: 1) species which had L:D ratios >1:1 in every vegetation association in which they occurred, 2 ) species which had L:D ratios <1:1 in the low elevation Coleoavne and/or Larrea-Lvcium- Gravia vegetation associations and >1:1 in all others, and 3 ) species which had L:D ratios (1:l in all but the Lycium-Gravia association (Table 5 and Figure8 2-4) .

Seventeen species { 5 7 % ) had L:D ratios L1:l in every vegetation association in which they occurred. Figures 2a-2c show the specific level of response that three of these species had.

Ten species had L:D ratios >1:1 in both the lowest and highest elevation associations but had L:D ratios <1:1 in at least one of the two vegetation associations that occur at intermediate points on the elevation/precipitation gradient (Tables 1 and 5). Figures 3a and 3b show two of the species that had this response to the drought.

Three species (13 % ) had L:D ratios >1:1 in only the Lvcium-Gravia vegetation association and L:D ratio <1:1 in all other vegetation associations. The two most common bunch grasses in the project area, 0. hvmenoides and S. sDeciosa had this response to the drought (Figures 4a and 4b).

DISCUSSION AND CONCLUSIONS

1. Vegetation associations present in the northern Mojave Desert do not respond similarly to prolonged drought (Tables 2 and 3 ) .

2. The effect that the prolonged drought had on the vegetation associations did not follow elevation and precipitation gradients (Tables 1, 2 and 3 ) . The high elevation (Coleocrvne and Lycium-Gravia) vegetation associations and to a lesser extent the lowest elevation association (Larrea-Ambrosia) were less affected by the drought. Vegetation associations at intermediate points on the elevation and precipitation gradient (i.e., Larrea-Lvcium-Gravia and low elevation Coleoavne), had many species that appeared to suffer substantial severe mortality during the drought (Table 3 ) .

3 .

4.

Numerous soil factors probably determine how drought affects both a vegetation association and an individual species. were highly variable between vegetation associations (Table 1 and personal observation), and this variation may help explain why individual species responded differently across vegetation associations. Additional research on plant-soil relationships, and how drought can affect species mortality, is necessary before definitive conclusions are possible.

Soil characteristics

Additional studies on how the L:D ratios present in the study plots sampled change as moisture conditions return towards the long-term average may help determine the dynamics of species populations in the northern Mojave Desert.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal Iiability or responsi- bility for the accuracy, completeness, or usefulness of any infomation, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Table 1. vegetation associations at Yucca Mountain.

General physiographic and abiotic characteristics of the five vegetation primary

Vegetation Association

Larrea- Ambrosia

Larrea-Lycium- Grayia

Low elevation Col eogyne

High Elevation Col eogyne

Elevation Range (m) 900-1050

1000-1200

1100-1300

1400-1700

Landform

Sandy alluvial plain

Young gravelly alluvial outwash

Old alluvial fans

Flat mountain tops and mesas

Relative Precipitation (1992 ave.)

Lowest (166 mm)

Intermediate (219 mm) Intermediate (212 mm)

Highest (260 mm)

Average Sot1 Depth (cm)

80+

60-100

15-45

30-45

Lyci urn-Grayi a 1150-1500 30-45 Ridge tops and Intermediate

sideslopes mountain (220 mm)

’ Personal observation of the authors.

Table 2. Total living and dead plants recorded in the vegetation associations at Yucca Mountain i n 1991.

Larrea Lycium Gra yi a

5,930

Larrea Ambrosia

Low High Lycium Elevation Elevation Grayia Col eogyne Coleogyne

4 , 882 2 , 512 14 , 316 Live 13 , 511

9 , 758 lants

Percent Alive I 58-1

4 , 129 I 5,411 I 661 I 3,139

58.9 I 47.4 I 79.2 I 82.0

Total

41,151

, 23,098

64.0

Table 3. Chi-square analysis of 1ive:dead ratios for 30 most common perennial species in the vegetation associations at Yucca Mountain in 1991.

Lycium Grayia

30 22.8

L:D >1:1

Total

87

L:D <1:1

Total

Larrea Ambrosia

Obs. 17 EXD. 15.9

O b s . 4 EXD. 5.1

21

VEGETATI4

Larrea Lycium Grayia

11 17.4

12 5.6

23

Obs. = number of species observed. Exp. = number of species expected.

IN ASSOCIATIONS

High Elevation Elevation Coleo ne

15.9 15.1

lg.1 1 s . 9

21 I 20 I--- X2 = 28.0 p<.OOl, df=4

Table 4. The total number of plants (alive and dead) identified by species in the study locations at Yucca Mountain in 1991.

SPECIES TOTAL LIVE

TOTAL DEAD

PERCENT ALIVE

Acamptopappus shockleyi* 2,099 482 81 Ambrosia dumosa* Atriplex canescens Atriplex confertifolia Ceratoides lanata Chrysothamnus nauseosus Chrysothamnus teretifolius Chrysothamnus viscidif1 orus Coleogyne ramosissima Encel la virginensis Ephedra nevadensi s Ephedra viridis Eriogonum fasiculatum Erioneurun pulchellum Grayia spinosa Gutierrezia sarothrae Haplopappus cooperi Haplopappus linearifolius Hilaria jamesii Hymenochlea salsola Krameria parvifolia Larrea triden ta ta Lyci um andersoni i Lycium pallidum Menodora spinescens Oryzopsis hymenoides Salazaria mexicana Sphaeralcea ambigua Stephanumeria pauciflora stipa speciosa

10,447 60

851 1,023

39 434 183

2,164 377

3 , 677 134

1,304 1,338 2,198 401

1,598 113

981 2,627 1,261 1,538 947

2,104 80

221 1,359

62 1,439

1,119

4,803 13

777 337

1 63 42

979 253 726 6

848 399 757 143 956 44 26

504 253 49

388 110

1,162 2,288

75 76 13

1,921

69 82 52 75 98 87 81 69 60 84 96 61 77 61 74 63 72 98 66 91 96 80 90 64 3

75 95 83 43

* The actual percent alive values for these species should be slightly lower. Some dead specimens in the Larrea-Ambrosia association could not be separated between these two species. Only those that were positively identifiable were used in this analysis.

Table 5 . Typical responses displayed by the 30 most common species analyzed for L:D ratios at Yucca Mountain. Nevada.

Response Type

1. Species that had L:D ratios 2 1:l in all of the vegetation associations in which they occurred.

2.

3 .

Species that occurred across the entire elevation/precipitation gradient and had L:D ratios 2 1:l in vegetation associations at the lowest and highest elevations, and L:D ratios < 1:l in vegetation associations at intermediate points on elevation/precipitation gradient.

Species that had L:D ratios <1:1 in all vegetation associations, except the Lycium- Grayia (i.e., highest elevation).

Species

Chrysothamnus viscidif1 ?rusl Chrysothamnus nauseosus- Chrysothamnus teretif oliusl Coleoyyne ramosissima Encelia virginensis Ephedra nevadenris Ephedra viridis- Gutierrezia sarothrael , Haplopappus 1inTarifolius- Hilaria jamesii- Krameria parvifolia Lycium andersonii Lycium pal 1 idum Larrea tridenta ta Salazaria mexicana Sphaeralcea ambigua Stephanomeria pauciflora Ambrosia dumosa Atriplex canescens Atriplex confertifolia Acamptopappus shockleyi Ceratoides lanata Erioneuron pulchellum Grayia spinosa Haplopappus cooperi Hymenoclea salsola Menodora spinescens

Eriogonum fasiculatum Oryzopsis hymenoides Stipa speciosa

f Species that were present only in the high elevation Coleogyne and Lycium-Grayia associations.

Figure Titles

Figure 1. a) Larrea-Ambrosia association, b) Larrea-Lycium-Grayia association, c) low elevation Coleogyne association, d) high elevation Coleogyne association, and e) the Lyciurn-Grayia association.

Figures 2a-2c. Three of the twelve species that had L:D rations 2 1:l in each of the vegetation associations in which they occurred, and the variation in species response between vegetation associations. Values above each set of bars are the L:D ratio for the species.

Figures 3a and 3b. Two of the eight species that had L:D ratios - > 1:l in the vegetation associations at the lowest and highest elevations, and L:D ratios 1:l in one or both of the vegetation associations that occurred at intermediate points on the elevation/precipitation gradient. Values above each set of bars are the L:D ratio for the species.

Figures 4a and 4b. Two of the three species that had L:D ratios - > 1:l in only the Lycium-Grayia vegetation association. Values above each set of bars are the L:D ratio for the species.

3

f

a. LYCIUM ANDERSON11

-I a 40

E 30 25

z 15

Lco3J

2 20

2 10 5 n

55 g so 5 45 7 16.9

I.

LA U G COLL COLH LG VEGETATION ASSOCIATION

LYCIUM PALLIDUM b.

so p 45 8 4Q

o 25

8 20

35

2 30 a

I 3 15 z

1L

5 I 5

0 LA U C COLL COLH LG

VEGETATION ASSOCIATION

LARREA TAIDENTATA C.

VEGETATION ASSOCIATION

Figure 2.

2 50

g 40 8 45 2 35 & 30 p: 25

3 10

z

20 1 15

I 5 0

LA LLG COLL COLH LG

a.

b.

Figure 3.

AMBROSIA DUMOSA

4.3 600 4

VEGETATION ASSOClATlON

GRAYIA SPINOSA

4.9 w

3 70

H p

840

f

g 10

0- 3 5 0

a 30

z 20 z U

0 LA LLG COLL COLH LG

VEGETATION ASSOCIATION

a.

b.

ORYZOPStS HYMENOIDES

200 180

160

140

120

100 80

60

40

20

0 LLG COLL COLH LG VEGETATION ASSOCIATION

STIPA SPECIOSA

1 zu 1.3

110

1 00 90 80 70

60

50

40 30 20 I O

0 LA LLG COLL COLH LG

VEGETATION ASSOCIATION

Figure 4.